From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 9861 invoked by alias); 3 Dec 2010 17:47:22 -0000 Received: (qmail 9839 invoked by uid 22791); 3 Dec 2010 17:47:20 -0000 X-SWARE-Spam-Status: No, hits=-1.1 required=5.0 tests=AWL,BAYES_00,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,FREEMAIL_ENVFROM_END_DIGIT,FREEMAIL_FROM,RCVD_IN_DNSWL_LOW X-Spam-Check-By: sourceware.org Received: from mail-fx0-f41.google.com (HELO mail-fx0-f41.google.com) (209.85.161.41) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Fri, 03 Dec 2010 17:47:12 +0000 Received: by fxm13 with SMTP id 13so8546057fxm.0 for ; Fri, 03 Dec 2010 09:47:10 -0800 (PST) Received: by 10.223.69.136 with SMTP id z8mr2389816fai.104.1291398429853; Fri, 03 Dec 2010 09:47:09 -0800 (PST) MIME-Version: 1.0 Received: by 10.223.117.193 with HTTP; Fri, 3 Dec 2010 09:46:49 -0800 (PST) In-Reply-To: References: <1291165591-12653-1-git-send-email-mattst88@gmail.com> From: Matt Turner Date: Fri, 03 Dec 2010 17:47:00 -0000 Message-ID: Subject: Re: [PATCH] MIPS: delete getpagesize.c To: "Joseph S. Myers" Cc: libc-ports@sourceware.org, "Maciej W. Rozycki" , Richard Henderson , Andrew Pinksi Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-IsSubscribed: yes Mailing-List: contact libc-ports-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Post: List-Help: , Sender: libc-ports-owner@sourceware.org X-SW-Source: 2010-12/txt/msg00002.txt.bz2 On Fri, Dec 3, 2010 at 5:21 PM, Joseph S. Myers w= rote: > On Tue, 30 Nov 2010, Matt Turner wrote: > >> MIPS' getpagesize.c just included ia64's getpagesize.c, which was in >> turn just the implementation sysdeps/unix/sysv/linux/getpagesize.c with >> ifdef __ASSUME_AT_PAGESIZE which is defined on all Linux platforms since >> 2.4.0-pre. Just use the generic implementation which will do the exact >> same thing. > > It looks like the default configuration for MIPS is for > arch_minimum_kernel=3D2.4.0 and __ASSUME_AT_PAGESIZE is only defined for > 2.4.1 or later, so this would change things for anyone configuring without > --enable-kernel. =A0Support for anything before 2.6 isn't exactly *useful* > with current glibc given that NPTL is the only threading implementation > available; should we increase arch_minimum_kernel for MIPS to 2.6.0 or > greater before putting this patch in? Looking at kernel-features.h, it seems that a huge number of syscalls were enabled first in 2.4.1. I don't have any problem at all with increasing arch_minimum_kernel, but I imagine some people do still use 2.4.x, so maybe we should just modify arch_minimum_kernel to be 2.4.1? I can't think of _any_ reason why someone would want to use plain 2.4.0. Matt